Search

The most common approaches for classification rely on the inference of a specific class. However, every category could be naturally organized within a taxonomic tree, from the most general concept to the specific element, and that is how human knowledge works. This representation avoids the necessity of learning roughly the same features for a range of very similar categories, and it is easier to understand and work with and provides a classification for each abstraction level. In this paper, we carry out an exhaustive study of different methods to perform multilevel classification applied to the task of classifying wild animals and plant species. Different convolutional backbones, data setups, and ensembling techniques are explored to find the model which provides the best performance. As our experimentation remarks, in order to achieve the best performance on the datasets that are arranged in a tree-like structure, the classifier must feature an EfficientNetB5 backbone with an input size of 300 × 300 px, followed by a multilevel classifier. In addition, a Multiscale Crop data augmentation process must be carried out. Finally, the accuracy of this setup is a 62% top-1 accuracy and 88% top-5 accuracy. The architecture could benefit for an accuracy boost if it is involved in an ensemble of cascade classifiers, but the computational demand is unbearable for any real application.

Cognitive architectures allow robots to perform their operations by drawing on a process that aims to simulate human reasoning. This paper presents an integrated semantic artificial memory system in cognitive architecture based on symbolic reasoning and a connective representation of the knowledge. This memory system attempts to simulate how humans learn to distinguish instances of particular objects within their class using a convolutional network to detect the relevant elements of an image. We use a vector with the extracted features to learn to discriminate an instance of another element from the same class. A novel feature of our approach is its autonomous learning process during the operation of the robot, integrating a deep learning embedding with a statistical classifier. The usefulness and robustness of this method are demonstrated by applying it to a social robot that learns to differentiate people. Finally, experiments are carried out to validate our approach, comparing the detection results with several alternative methods. This work has been funded by the Spanish Government TIN2016-76515-R grant for the COMBAHO project, supported with Feder funds. This work has also been supported by a Spanish grant for PhD studies ACIF/2017/243 and FPU16/00887.

In this paper, we present a novel deep learning-based architecture, which is under the scope of expert and intelligent systems, to perform accurate real-time tridimensional hand pose estimation using a single RGB frame as an input, so there is no need to use multiple cameras or points of view, or RGB-D devices. The proposed pipeline is composed of two convolutional neural network architectures. The first one is in charge of detecting the hand in the image. The second one is able to accurately infer the tridimensional position of the joints retrieving, thus, the full hand pose. To do this, we captured our own large-scale dataset composed of images of hands and the corresponding 3D joints annotations. The proposal achieved a 3D hand pose mean error of below 5 mm on both the proposed dataset and Stereo Hand Pose Tracking Benchmark, which is a public dataset. Our method also outperforms the state-of-the-art methods. We also demonstrate in this paper the application of the proposal to perform a robotic hand teleoperation with high success. This work has been supported by the Spanish Government TIN2016-76515R Grant, supported with Feder funds. This work has also been supported by a Spanish grant for PhD studies ACIF/2017/243

Twenty-four studies on twenty-three distinct languages and eleven social media illustrate the steady interest in deep learning approaches for multilingual sentiment analysis of social media. We improve over previous reviews with wider coverage from 2017 to 2020 as well as a study focused on the underlying ideas and commonalities behind the different solutions to achieve multilingual sentiment analysis. Interesting findings of our research are (i) the shift of research interest to cross-lingual and code-switching approaches, (ii) the apparent stagnation of the less complex architectures derived from a backbone featuring an embedding layer, a feature extractor based on a single CNN or LSTM and a classifier, (iii) the lack of approaches tackling multilingual aspect-based sentiment analysis through deep learning, and, surprisingly, (iv) the lack of more complex architectures such as the transformers-based, despite results suggest the more difficult tasks requires more elaborated architectures. This research work has been partially funded by the Generalitat Valenciana (Conselleria d’Educació, Investigació, Cultura i Esport) and the Spanish Government through the projects SIIA (PROMETEO/2018/089, PROMETEU/2018/089) and LIVING-LANG (RTI2018-094653-B-C22).

In this work, we introduce HaReS, a hand rehabilitation system. Our proposal integrates a series of exercises, jointly developed with a foundation for those with motor and cognitive injuries, that are aimed at improving the skills of patients and the adherence to the rehabilitation plan. Our system takes advantage of a low-cost hand-tracking device to provide a quantitative analysis of the performance of the patient. It also integrates a low-cost surface electromyography (sEMG) sensor in order to provide insight about which muscles are being activated while completing the exercises. It is also modular and can be deployed on a social robot. We tested our proposal in two different facilities for rehabilitation with high success. The therapists and patients felt more motivation while using HaReS, which improved the adherence to the rehabilitation plan. In addition, the therapists were able to provide services to more patients than when they used their traditional methodology. This work was funded by a Spanish Government PID2019-104818RB-I00 grant, supported by Feder funds. It was also supported by Spanish grants for PhD studies ACIF/2017/243 and FPU16/00887.

Public Sector Information is considered to play a fundamental role in the growth of the knowledge economy and improvements in society. Given the difficulty in publishing and maintaining all available data, due to budget constraints, institutions need to select which data to publish, giving priority to data most likely to generate social and economic impact. Priority of publication could become an even more significant problem in Smart Cities: as huge amounts of information are generated from different domains, the way data is prioritized and thus reused, could be a determining factor in promoting, among others, new and sustainable business opportunities for local entrepreneurs, and to improve citizen quality of life. However, people in charge of prioritizing which data to publish through open data portals (such as Chief Data Officers, or CDOs) do not have available any specific support in their decision-making process. In this work, a proposal of a framework for prioritization of open data publication as well as its application to Smart Cities is presented. This specific application of the framework relies on OSS (Open Source Software) indicators to help making decisions on the most relevant data to publish focused on developers and businesses operating within the Smart City context. This work was funded by (i) Ministerio de Economía e Innovación (Spain) TIN2015-69957-R (MINECO/ERDF, EU) project and TIN2016-78103-C2-2-R (MINECO/ERDF, EU) project, (ii) POCTEP 4IE project (0045-4IE-4-P), and (iii) Consejería de Economía e Infraestructuras/Junta de Extremadura (Spain) - European Regional Development Fund (ERDF)- GR18112 project and IB16055 project.

In this paper, we present a robotic workcell for task automation in footwear manufacturing such as sole digitization, glue dispensing, and sole manipulation from different places within the factory plant. We aim to make progress towards shoe industry 4.0. To achieve it, we have implemented a novel sole grasping method, compatible with soles of different shapes, sizes, and materials, by exploiting the particular characteristics of these objects. Our proposal is able to work well with low density point clouds from a single RGBD camera and also with dense point clouds obtained from a laser scanner digitizer. The method computes antipodal grasping points from visual data in both cases and it does not require a previous recognition of sole. It relies on sole contour extraction using concave hulls and measuring the curvature on contour areas. Our method was tested both in a simulated environment and in real conditions of manufacturing at INESCOP facilities, processing 20 soles with different sizes and characteristics. Grasps were performed in two different configurations, obtaining an average score of 97.5% of successful real grasps for soles without heel made with materials of low or medium flexibility. In both cases, the grasping method was tested without carrying out tactile control throughout the task. Research work was completely funded by the European Commission and FEDER through the COMMANDIA project (SOE2/P1/F0638), supported by Interreg-V Sudoe. Part of the facilities used were provided by the Footwear Technological Institute (INESCOP).

Dementia is a syndrome that is characterised by the decline of different cognitive abilities. A high rate of deaths and high cost for detection, treatments, and patients care count amongst its consequences. Although there is no cure for dementia, a timely diagnosis helps in obtaining necessary support, appropriate medication, and maintenance, as far as possible, of engagement in intellectual, social, and physical activities. The early detection of Alzheimer Disease (AD) is considered to be of high importance for improving the quality of life of patients and their families. In particular, Virtual Reality (VR) is an expanding tool that can be used in order to assess cognitive abilities while navigating through a Virtual Environment (VE). The paper summarises common AD screening and diagnosis techniques focusing on the latest approaches that are based on Virtual Environments, behaviour analysis, and emotions recognition, aiming to provide more reliable and non-invasive diagnostics at home or in a clinical environment. Furthermore, different AD diagnosis evaluation methods and metrics are presented and discussed together with an overview of the different datasets. This work has been funded by the Spanish Government PID2019-104818RB-I00 grant for the MoDeaAS project, supported with Feder funds.

Over one billion people in the world live with some form of disability. This is incessantly increasing due to aging population and chronic diseases. Among the emerging social needs, rehabilitation services are the most required. However, they are scarce and expensive what considerably limits access to them. In this paper, we propose EVA, an augmented reality platform to engage and supervise rehabilitation sessions at home using low-cost sensors. It also stores the user’s statistics and allows therapists to tailor the exercise programs according to their performance. This system has been evaluated in both qualitative and quantitative ways obtaining very promising results. This work has been supported by the Spanish Government TIN2016-76515R Grant, supported with Feder funds. Edmanuel Cruz is funded by a Panamenian grant for Ph.D. studies IFARHU and SENACYT 270-2016-207. This work has also been supported by a Spanish grant for PhD studies ACIF/2017/243 and FPU16/00887. Thanks also to Nvidia for the generous donation of a Titan Xp and a Quadro P6000.

The goal of this paper is to present a robotic workcell to automate several tasks of the cementing process in footwear manufacturing. Our cell's main applications are sole digitization of a wide variety of footwear, glue dispensing and sole grasping from conveyor belts. This cell is made up of a manipulator arm endowed with a gripper, a conveyor belt and a 3D scanner. We have integrated all the elements into a ROS simulation environment facilitating control and communication among them, also providing flexibility to support future extensions. We propose a novel method to grasp soles of different shape, size and material, exploiting the particular characteristics of these objects. Our method relies on object contour extraction using concave hulls. We evaluate it on point clouds of 16 digitized real soles in three different scenarios: concave hull, k-NNs extension and PCA correction. While we have tested this workcell in a simulated environment, the presented system's performance is scheduled to be tested on a real setup at INESCOP facilities in the upcoming months. Work funded by the European Commission and FEDER funds through the COMMANDIA project (SOE2/P1/F0638), supported by Interreg-V Sudoe.

This work presents an end-to-end method based on deep neural networks for audio-to-score music transcription of monophonic excerpts. Unlike existing music transcription methods, which normally perform pitch estimation, the proposed approach is formulated as an end-to-end task that outputs a notation-level music score. Using an audio file as input, modeled as a sequence of frames, a deep neural network is trained to provide a sequence of music symbols encoding a score, including key and time signatures, barlines, notes (with their pitch spelling and duration) and rests. Our framework is based on a Convolutional Recurrent Neural Network (CRNN) with Connectionist Temporal Classification (CTC) loss function trained in an end-to-end fashion, without requiring to align the input frames with the output symbols. A total of 246,870 incipits from the Répertoire International des Sources Musicales online catalog were synthesized using different timbres and tempos to build the training data. Alternative input representations (raw audio, Short-Time Fourier Transform (STFT), log-spaced STFT and Constant-Q transform) were evaluated for this task, as well as different output representations (Plaine & Easie Code, Kern, and a purpose-designed output). Results show that it is feasible to directly infer score representations from audio files and most errors come from music notation ambiguities and metering (time signatures and barlines). This work has been supported by the Spanish "Ministerio de Ciencia e Innovación" through Project HISPAMUS (No. TIN2017-86576-R supported by EU FEDER funds).

Tactile perception is a rich source of information for robotic grasping: it allows a robot to identify a grasped object and assess the stability of a grasp, among other things. However, the tactile sensor must come into contact with the target object in order to produce readings. As a result, tactile data can only be attained if a real contact is made. We propose to overcome this restriction by employing a method that models the behaviour of a tactile sensor using 3D vision and grasp information as a stimulus. Our system regresses the quantified tactile response that would be experienced if this grasp were performed on the object. We experiment with 16 items and 4 tactile data modalities to show that our proposal learns this task with low error. This work was supported in part by the Spanish Government and the FEDER Funds (BES-2016-078290, PRX19/00289, RTI2018-094279-B-100) and in part by the European Commission (COMMANDIA SOE2/P1/F0638), action supported by Interreg-V Sudoe.

Deep learning-based methods have proven to be the best performers when it comes to object recognition cues both in images and tridimensional data. Nonetheless, when it comes to 3D object recognition, the authors tend to convert the 3D data to images and then perform their classification. However, despite its accuracy, this approach has some issues. In this work, we present a deep learning pipeline for object recognition that takes a point cloud as input and provides the classification probabilities as output. Our proposal is trained on synthetic CAD objects and is able to perform accurately when fed with real data provided by commercial sensors. Unlike most approaches, our method is specifically trained to work on partial views of the objects rather than on a full representation, which is not the representation of the objects as captured by commercial sensors. We trained our proposal with the ModelNet10 dataset and achieved a 78.39% accuracy. We also tested it by adding noise to the dataset and against a number of datasets and real data with high success. This work has been funded by the Spanish Government TIN2016-76515-R grant for the COMBAHO project, supported with Feder funds. It has also been supported by Spanish grants for PhD studies ACIF/2017/243 and FPU16/00887.

Grasping and manipulating objects with robotic hands depend largely on the features of the object to be used. Especially, features such as softness and deformability are crucial to take into account during the manipulation tasks. Indeed, positions of the fingers and forces to be applied by the robot hand when manipulating an object must be adapted to the caused deformation. For unknown objects, a previous recognition stage is usually needed to get the features of the object, and the manipulation strategies must be adapted depending on that recognition stage. To obtain a precise control in the manipulation task, a complex object model is usually needed and performed, for example using the Finite Element Method. However, these models require a complete discretization of the object and they are time-consuming for the performance of the manipulation tasks. For that reason, in this paper a new control strategy, based on a minimal spring model of the objects, is presented and used for the control of the robot hand. This paper also presents an adaptable tactile-servo control scheme that can be used in in-hand manipulation tasks of deformable objects. Tactile control is based on achieving and maintaining a force value at the contact points which changes according to the object softness, a feature estimated in an initial recognition stage. Research supported by Spanish Ministry of Economy, European FEDER funds, the Valencia Regional Government and University of Alicante, through projects DPI2012-32390, DPI2015-68087-R, PROMETEO/2013/085 and GRE 15-05.

The capacity of a robot to automatically adapt to new environments is crucial, especially in social robotics. Often, when these robots are deployed in home or office environments, they tend to fail because they lack the ability to adapt to new and continuously changing scenarios. In order to accomplish this task, robots must obtain new information from the environment, and then add it to their already learned knowledge. Deep learning techniques are often used to tackle this problem successfully. However, these approaches, complete retraining of the models, which is highly time-consuming. In this work, several strategies are tested to find the best way to include new knowledge in an already learned model in a deep learning pipeline, putting the spotlight on the time spent for this training. We tackle the localization problem in the long term with a deep learning approach and testing several retraining strategies. The results of the experiments are discussed and, finally, the best approach is deployed on a Pepper robot. The capacity of a robot to automatically adapt to new environments is crucial, especially in social robotics. Often, when these robots are deployed in home or office environments, they tend to fail because they lack the ability to adapt to new and continuously changing scenarios. In order to accomplish this task, robots must obtain new information from the environment, and then add it to their already learned knowledge. Deep learning techniques are often used to tackle this problem successfully. However, these approaches, complete retraining of the models, which is highly time-consuming. In this work, several strategies are tested to find the best way to include new knowledge in an already learned model in a deep learning pipeline, putting the spotlight on the time spent for this training. We tackle the localization problem in the long term with a deep learning approach and testing several retraining strategies. The results of the experiments are discussed and, finally, the best approach is deployed on a Pepper robot.

As estimated by the World Health Organization, there are millions of people who lives with some form of vision impairment. As a consequence, some of them present mobility problems in outdoor environments. With the aim of helping them, we propose in this work a system which is capable of delivering the position of potential obstacles in outdoor scenarios. Our approach is based on non-intrusive wearable devices and focuses also on being low-cost. First, a depth map of the scene is estimated from a color image, which provides 3D information of the environment. Then, an urban object detector is in charge of detecting the semantics of the objects in the scene. Finally, the three-dimensional and semantic data is summarized in a simpler representation of the potential obstacles the users have in front of them. This information is transmitted to the user through spoken or haptic feedback. Our system is able to run at about 3.8 fps and achieved a 87.99% mean accuracy in obstacle presence detection. Finally, we deployed our system in a pilot test which involved an actual person with vision impairment, who validated the effectiveness of our proposal for improving its navigation capabilities in outdoors. This work has been supported by the Spanish Government TIN2016-76515R Grant, supported with Feder funds, the University of Alicante project GRE16-19, and by the Valencian Government project GV/2018/022. Edmanuel Cruz is funded by a Panamenian grant for PhD studies IFARHU & SENACYT 270-2016-207. This work has also been supported by a Spanish grant for PhD studies ACIF/2017/243. Thanks also to Nvidia for the generous donation of a Titan Xp and a Quadro P6000.

Large datasets computing is a research problem as well as a huge challenge due to massive amounts of data that are mined and crunched in order to successfully analyze these massive datasets because they constitute a valuable source of information over different and cross-folded domains, and therefore it represents an irreplaceable opportunity. Hence, the increasing number of environments that use data-intensive computations need more complex calculations than the ones applied to grid-based infrastructures. In this way, this paper analyzes the most commonly used algorithms regarding to this complex problem of handling large datasets whose part of research efforts are focused on reducing dimensional space. Consequently, we present a novel machine learning method that reduces dimensional space in large datasets. This approach is carried out by developing different phases: merging all datasets as a huge one, performing the Extract, Transform and Load (ETL) process, applying the Principal Component Analysis (PCA) algorithm to machine learning techniques, and finally displaying the data results by means of dashboards. The major contribution in this paper is the development of a novel architecture divided into five phases that presents an hybrid method of machine learning for reducing dimensional space in large datasets. In order to verify the correctness of our proposal, we have presented a case study with a complex dataset, specifically an epileptic seizure recognition database. The experiments carried out are very promising since they present very encouraging results to be applied to a great number of different domains. This work was partially funded by Grant RTI2018-094283-B-C32, ECLIPSE-UA (Spanish Ministry of Education and Science), and in part by the Lucentia AGI Grant. This work was partially funded by GENDER-NET Plus Joint Call on Gender an UN Sustainable Development Goals (European Commission - Grant Agreement 741874), funded in Spain by “La Caixa” Foundation (ID 100010434) with code LCF/PR/DE18/52010001 to MTH.

International audience; The goal of this paper is to predict 3D object shape to improve the visual perception of robots in grasping and manipulation tasks. The planning of image-based robotic manipulation tasks depends on the recognition of the object's shape. Mostly, the manipulator robots usually use a camera with configuration eye-in-hand. This fact limits the calculation of the grip on the visible part of the object. In this paper, we present a 3D Deep Convolutional Neural Network to predict the hidden parts of objects from a single-view and to accomplish recovering the complete shape of them. We have tested our proposal with both previously seen objects and novel objects from a well-known dataset.

In several large retail stores, such as malls, sport or food stores, the customer often feels lost due to the difficulty in finding a product. Although these large stores usually have visual signs to guide customers toward specific products, sometimes these signs are also hard to find and are not updated. In this paper, we propose a system that jointly combines deep learning and augmented reality techniques to provide the customer with useful information. First, the proposed system learns the visual appearance of different areas in the store using a deep learning architecture. Then, customers can use their mobile devices to take a picture of the area where they are located within the store. Uploading this image to the system trained for image classification, we are able to identify the area where the customer is located. Then, using this information and novel augmented reality techniques, we provide information about the area where the customer is located: route to another area where a product is available, 3D product visualization, user location, analytics, etc. The system developed is able to successfully locate a user in an example store with 98% accuracy. The combination of deep learning systems together with augmented reality techniques shows promising results toward improving user experience in retail/commerce applications: branding, advance visualization, personalization, enhanced customer experience, etc. In several large retail stores, such as malls, sport or food stores, the customer often feels lost due to the difficulty in finding a product. Although these large stores usually have visual signs to guide customers toward specific products, sometimes these signs are also hard to find and are not updated. In this paper, we propose a system that jointly combines deep learning and augmented reality techniques to provide the customer with useful information. First, the proposed system learns the visual appearance of different areas in the store using a deep learning architecture. Then, customers can use their mobile devices to take a picture of the area where they are located within the store. Uploading this image to the system trained for image classification, we are able to identify the area where the customer is located. Then, using this information and novel augmented reality techniques, we provide information about the area where the customer is located: route to another area where a product is available, 3D product visualization, user location, analytics, etc. The system developed is able to successfully locate a user in an example store with 98% accuracy. The combination of deep learning systems together with augmented reality techniques shows promising results toward improving user experience in retail/commerce applications: branding, advance visualization, personalization, enhanced customer experience, etc.

Data mining has proven to be very useful in order to extract information from data in many different contexts. However, due to the complexity of data mining techniques, it is required the know-how of an expert in this field to select and use them. Actually, adequately applying data mining is out of the reach of novice users which have expertise in their area of work, but lack skills to employ these techniques. In this paper, we use both model-driven engineering and scientific workflow standards and tools in order to develop named S3Mining framework, which supports novice users in the process of selecting the data mining classification algorithm that better fits with their data and goal. To this aim, this selection process uses the past experiences of expert data miners with the application of classification techniques over their own datasets. The contributions of our S3Mining framework are as follows: (i) an approach to create a knowledge base which stores the past experiences of experts users, (ii) a process that provides the expert users with utilities for the construction of classifiers’ recommenders based on the existing knowledge base, (iii) a system that allows novice data miners to use these recommenders for discovering the classifiers that better fit for solving their problem at hand, and (iv) a public implementation of the framework’s workflows. Finally, an experimental evaluation has been conducted to shown the feasibility of our framework. This work has been partially funded by Spanish Government through the research projects TIN2017-86520-C3-3-R and TIN2016-78103-C2-2-R.

Any robot should be provided with a proper representation of its environment in order to perform navigation and other tasks. In addition to metrical approaches, topological mapping generates graph representations in which nodes and edges correspond to locations and transitions. In this article, we present LexToMap, a topological mapping procedure that relies on image annotations. These annotations, represented in this work by lexical labels, are obtained from pre-trained deep learning models, namely CNNs, and are used to estimate image similarities. Moreover, the lexical labels contribute to the descriptive capabilities of the topological maps. The proposal has been evaluated using the KTH-IDOL 2 data-set, which consists of image sequences acquired within an indoor environment under three different lighting conditions. The generality of the procedure as well as the descriptive capabilities of the generated maps validate the proposal. Any robot should be provided with a proper representation of its environment in order to perform navigation and other tasks. In addition to metrical approaches, topological mapping generates graph representations in which nodes and edges correspond to locations and transitions. In this article, we present LexToMap, a topological mapping procedure that relies on image annotations. These annotations, represented in this work by lexical labels, are obtained from pre-trained deep learning models, namely CNNs, and are used to estimate image similarities. Moreover, the lexical labels contribute to the descriptive capabilities of the topological maps. The proposal has been evaluated using the KTH-IDOL 2 data-set, which consists of image sequences acquired within an indoor environment under three different lighting conditions. The generality of the procedure as well as the descriptive capabilities of the generated maps validate the proposal.

Schaeffer's sign language consists of a reduced set of gestures designed to help children with autism or cognitive learning disabilities to develop adequate communication skills. Our automatic recognition system for Schaeffer's gesture language uses the information provided by an RGB-D camera to capture body motion and recognize gestures using dynamic time warping combined with k-nearest neighbors methods. The learning process is reinforced by the interaction with the proposed system that accelerates learning itself thus helping both children and educators. To demonstrate the validity of the system, a set of qualitative experiments with children were carried out. As a result, a system which is able to recognize a subset of 11 gestures of Schaeffer's sign language online was achieved. This work has been supported by the Spanish Government DPI2013-40534-R Grant, supported with Feder funds.

In this work, the authors propose several techniques for accelerating a modern action recognition pipeline. This article reviewed several recent and popular action recognition works and selected two of them as part of the tools used for improving the aforementioned acceleration. Specifically, temporal segment networks (TSN), a convolutional neural network (CNN) framework that makes use of a small number of video frames for obtaining robust predictions which have allowed to win the first place in the 2016 ActivityNet challenge, and MotionNet, a convolutional-transposed CNN that is capable of inferring optical flow RGB frames. Together with the last proposal, this article integrated a new software for decoding videos that takes advantage of NVIDIA GPUs. This article shows a proof of concept for this approach by training the RGB stream of the TSN network in videos loaded with NVIDIA Video Loader (NVVL) of a subset of daily actions from the University of Central Florida 101 dataset.

In the past, methods for hand sign recognition have been successfully tested in Human Robot Interaction (HRI) using traditional methodologies based on static image features and machine learning. However, the recognition of gestures in video sequences is a problem still open, because current detection methods achieve low scores when the background is undefined or in unstructured scenarios. Deep learning techniques are being applied to approach a solution for this problem in recent years. In this paper, we present a study in which we analyse the performance of a 3DCNN architecture for hand gesture recognition in an unstructured scenario. The system yields a score of 73% in both accuracy and F1. The aim of the work is the implementation of a system for commanding robots with gestures recorded by video in real scenarios. This work was funded by the Ministry of Economy, Industry and Competitiveness from the Spanish Government through the DPI2015-68087-R and the pre-doctoral grant BES-2016-078290, by the European Commission and FEDER funds through the project COMMANDIA (SOE2/P1/F0638), action supported by Interreg-V Sudoe.

As it is well known, some versions of the Pepper robot provide poor depth perception due to the lenses it has in front of the tridimensional sensor. In this paper, we present a method to improving that faulty 3D perception. Our proposal is based on a combination of the actual depth readings of Pepper and a deep learning-based monocular depth estimation. As shown, the combination of both of them provides a better 3D representation of the scene. In previous works we made an initial approximation of this fusion technique, but it had some drawbacks. In this paper we analyze the pros and cons of the Pepper readings, the monocular depth estimation method and our previous fusion method. Finally, we demonstrate that the proposed fusion method outperforms them all. This work was supported in part by the Spanish Government, through Feder funds under Grant TIN2016-76515R, and in part by the Spanish Grant for Ph.D. studies under Grant ACIF/2017/243 and Grant FPU16/00887. The work of E. Cruz was supported by the Panamenian Grant for Ph.D. studies IFARHU & SENACYT under Grant 270-2016-207.

The abundance of digital media information coming from different sources, completely redefines approaches to media content production management and distribution for all contexts (i.e. technical, business and operational). Such content includes descriptive information (i.e. metadata) about an asset (e.g. a movie, song or game), as well as playable media (e.g. audio or video files). Metadata is organised following a variety of inconsistent structures and formats that are supplied by various content providers. Some challenges have been addressed in terms of standardising and enriching media assets metadata from a semantic perspective. Well known examples include Europeana and DBpedia. Nevertheless, due to the ongoing variability and evolution of digital contents, constant support and creation of new semantic representations are necessary. This article presents an ontology schema covering the requirements of users (content providers and content consumers) involved in the overall life cycle of a digital media asset, which has been designed and developed for a real scenario. The construction of this schema has been documented and evaluated following a methodology supported by quantitative and qualitative metrics. As part of the tangible results, the following outcomes were produced: (i) an RDF/XML schema available via Zenodo and GitHub; (ii) competence questions used for validation are published at GitHub; (iii) an exemplary ontology repository; and (iv) CRUD (Create, Read, Update and Delete) technologies for managing semantic repositories based on such schema. These results form an active part of the framework of a European project and other ongoing research initiatives. This research work has been partially funded by the European Commission under the 7th Framework Programme for Research and Technological Development, the Spanish Government, the Generalitat Valenciana and the University of Alicante thought the projects: SAM(FP7-611312); SIIA(PROMETEU/2018/089); TIN2015-65136-C2-2-R; RTI2018-094653-B-C22; RTI2018-094649-B-I00; GRE16-01; and Social-Univ 2.0 (ENCARGOINTERNO5-19EN).

Ambient assisted living (AAL) environments are currently a key focus of interest as an option to assist and monitor disabled and elderly people. These systems can improve their quality of life and personal autonomy by detecting events such as entering potentially dangerous areas, potential fall events, or extended stays in the same place. Nonetheless, there are areas that remain outside the scope of AAL systems due to the placement of cameras. There also exist sources of danger in the scope of the camera that the AAL system cannot detect. These sources of danger are relatively small in size, occluded, or nonstatic. To solve this problem, we propose the inclusion of a robot which maps such uncovered areas looking for new potentially dangerous areas that go unnoticed by the AAL. The robot then sends this information to the AAL system in order to improve its performance. Experimentation in real-life scenarios successfully validates our approach. This work was supported by the Spanish Government TIN2016-76515R Grant, supported with FEDER funds. This work was also supported by a Spanish grant for PhD studies (ACIF/2017/243 and FPU16/00887).

In this study, we use unmanned aerial vehicles equipped with multispectral cameras to search for bodies in maritime rescue operations. A series of flights were performed in open‐water scenarios in the northwest of Spain, using a certified aquatic rescue dummy in dangerous areas and real people when the weather conditions allowed it. The multispectral images were aligned and used to train a convolutional neural network for body detection. An exhaustive evaluation was performed to assess the best combination of spectral channels for this task. Three approaches based on a MobileNet topology were evaluated, using (a) the full image, (b) a sliding window, and (c) a precise localization method. The first method classifies an input image as containing a body or not, the second uses a sliding window to yield a class for each subimage, and the third uses transposed convolutions returning a binary output in which the body pixels are marked. In all cases, the MobileNet architecture was modified by adding custom layers and preprocessing the input to align the multispectral camera channels. Evaluation shows that the proposed methods yield reliable results, obtaining the best classification performance when combining green, red‐edge, and near‐infrared channels. We conclude that the precise localization approach is the most suitable method, obtaining a similar accuracy as the sliding window but achieving a spatial localization close to 1 m. The presented system is about to be implemented for real maritime rescue operations carried out by Babcock Mission Critical Services Spain. This study was performed in collaboration with BabcockMCS Spain and funded by the Galicia Region Government through the Civil UAVs Initiative program, the Spanish Government’s Ministry of Economy, Industry, and Competitiveness through the RTC‐2014‐1863‐8 and INAER4‐14Y (IDI‐20141234) projects, and the grant number 730897 under the HPC‐EUROPA3 project supported by Horizon 2020.

This Special Issue is focused on breakthrough developments in the field of biosensors and current scientific progress in biomedical signal processing. The papers address innovative solutions in assistance robotics based on bioelectrical signals, including: Affordable biosensor technology, affordable assistive-robotics devices, new techniques in myoelectric control and advances in brain–machine interfacing.

This paper presents a system that combines computer vision and surface electromyography techniques to perform grasping tasks with a robotic hand. In order to achieve a reliable grasping action, the vision-driven system is used to compute pre-grasping poses of the robotic system based on the analysis of tridimensional object features. Then, the human operator can correct the pre-grasping pose of the robot using surface electromyographic signals from the forearm during wrist flexion and extension. Weak wrist flexions and extensions allow a fine adjustment of the robotic system to grasp the object and finally, when the operator considers that the grasping position is optimal, a strong flexion is performed to initiate the grasping of the object. The system has been tested with several subjects to check its performance showing a grasping accuracy of around 95% of the attempted grasps which increases in more than a 13% the grasping accuracy of previous experiments in which electromyographic control was not implemented. This work was funded by the Spanish Government’s Ministry of Economy, Industry and Competitiveness through the DPI2015-68087-R, by the European Commission’s and FEDER funds through the COMMANDIA (SOE2/P1/F0638) action supported by Interreg-V Sudoe and by University of Alicante through project GRE16-20, Control Platform for a Robotic Hand based on Electromyographic Signals.

3D data sensors provide an enormous amount of information. It is necessary to develop efficient methods to manage this information under certain time, bandwidth or storage space requirements. In this work, we propose a 3D compression and decompression method. This method also allows the use of the compressed data for a registration process. First, points are selected and grouped, using a 3D-model based on planar surfaces. Next, we use a fast variant of Gaussian Mixture Models and an Expectation-Maximization algorithm to replace the points grouped in the previous step with a set of Gaussian distributions. These learned models can be used as features to find matches between two consecutive poses and apply 3D pose registration using RANSAC. Finally, the 3D map can be obtained by decompressing the models. This work has been supported by the Spanish Government TIN2016-76515-R Grant, supported with Feder funds.

Object recognition in 3D scenes is a research field in which there is intense activity guided by the problems related to the use of 3D point clouds. Some of these problems are influenced by the presence of noise in the cloud that reduces the effectiveness of a recognition process. This work proposes a method for dealing with the noise present in point clouds by applying the growing neural gas (GNG) network filtering algorithm. This method is able to represent the input data with the desired number of neurons while preserving the topology of the input space. The GNG obtained results which were compared with a Voxel grid filter to determine the efficacy of our approach. Moreover, since a stage of the recognition process includes the detection of keypoints in a cloud, we evaluated different keypoint detectors to determine which one produces the best results in the selected pipeline. Experiments show how the GNG method yields better recognition results than other filtering algorithms when noise is present. Object recognition in 3D scenes is a research field in which there is intense activity guided by the problems related to the use of 3D point clouds. Some of these problems are influenced by the presence of noise in the cloud that reduces the effectiveness of a recognition process. This work proposes a method for dealing with the noise present in point clouds by applying the growing neural gas (GNG) network filtering algorithm. This method is able to represent the input data with the desired number of neurons while preserving the topology of the input space. The GNG obtained results which were compared with a Voxel grid filter to determine the efficacy of our approach. Moreover, since a stage of the recognition process includes the detection of keypoints in a cloud, we evaluated different keypoint detectors to determine which one produces the best results in the selected pipeline. Experiments show how the GNG method yields better recognition results than other filtering algorithms when noise is present.

Despite the outstanding results of Convolutional Neural Networks (CNNs) in object recognition and classification, there are still some open problems to address when applying these solutions to real-world problems. Specifically, CNNs struggle to generalize under challenging scenarios, like recognizing the variability and heterogeneity of the instances of elements belonging to the same category. Some of these difficulties are directly related to the input information, 2D-based methods still show a lack of robustness against strong lighting variations, for example. In this paper, we propose to merge techniques using both 2D and 3D information to overcome these problems. Specifically, we take advantage of the spatial information in the 3D data to segment objects in the image and build an object classifier, and the classification capabilities of CNNs to semi-supervisedly label each object image for training. As the experimental results demonstrate, our model can successfully generalize for categories with high intra-class variability and outperform the accuracy of a well-known CNN model. Despite the outstanding results of Convolutional Neural Networks (CNNs) in object recognition and classification, there are still some open problems to address when applying these solutions to real-world problems. Specifically, CNNs struggle to generalize under challenging scenarios, like recognizing the variability and heterogeneity of the instances of elements belonging to the same category. Some of these difficulties are directly related to the input information, 2D-based methods still show a lack of robustness against strong lighting variations, for example. In this paper, we propose to merge techniques using both 2D and 3D information to overcome these problems. Specifically, we take advantage of the spatial information in the 3D data to segment objects in the image and build an object classifier, and the classification capabilities of CNNs to semi-supervisedly label each object image for training. As the experimental results demonstrate, our model can successfully generalize for categories with high intra-class variability and outperform the accuracy of a well-known CNN model.

This work presents a method for oil-spill detection on Spanish coasts using aerial Side-Looking Airborne Radar (SLAR) images, which are captured using a Terma sensor. The proposed method uses grayscale image processing techniques to identify the dark spots that represent oil slicks on the sea. The approach is based on two steps. First, the noise regions caused by aircraft movements are detected and labeled in order to avoid the detection of false-positives. Second, a segmentation process guided by a map saliency technique is used to detect image regions that represent oil slicks. The results show that the proposed method is an improvement on the previous approaches for this task when employing SLAR images. This work was supported by the Spanish Ministry of Economy and Competitiveness through the ONTIME research project (RTC-2014-1863-8).

In this work, we carry out a study of the effect of adverse conditions, which characterize real-world scenes, on the accuracy of a Convolutional Neural Network applied to 3D object class recognition. Firstly, we discuss possible ways of representing 3D data to feed the network. In addition, we propose a set of representations to be tested. Those representations consist of a grid-like structure (fixed and adaptive) and a measure for the occupancy of each cell of the grid (binary and normalized point density). After that, we propose and implement a Convolutional Neural Network for 3D object recognition using Caffe. At last, we carry out an in-depth study of the performance of the network over a 3D CAD model dataset, the Princeton ModelNet project, synthetically simulating occlusions and noise models featured by common RGB-D sensors. The results show that the volumetric representations for 3D data play a key role on the recognition process and Convolutional Neural Network can be considerably robust to noise and occlusions if a proper representation is chosen. This work has been supported by the Spanish Government DPI2013-40534-R grant for the SIRMAVED project, also supported with FEDER funds. This work has also been funded by the grant “Ayudas para Estudios de Máster e Iniciación a la Investigación” from the University of Alicante.

Synthetic Aperture Radar (SAR) polarimetry is an active and fruitful field of research in Earth observation. Polarimetry provides sensitivity to the soil moisture, as well as to the structural and geometric properties of the targets under observation, allowing a more accurate identification and classification than with non-polarimetric data. Moreover, the increasing number of spaceborne SAR systems equipped with polarimetric capabilities, as well as future planned missions, enables the advance in this research field at all levels, from theory and physical modeling to final applications.

Wireless sensor networks are used to sample the environment in a distributed way. Therefore, it is mandatory for all of the measurements to be tightly synchronized in order to guarantee that every sensor is sampling the environment at the exact same instant of time. The synchronization drift gets bigger in environments suffering from temperature variations. Thus, this work is focused on improving time synchronization under deployments with temperature variations. The working hypothesis demonstrated in this work is that the clock skew of two nodes (the ratio of the real frequencies of the oscillators) is composed of a multiplicative combination of two main components: the clock skew due to the variations between the cut of the crystal of each oscillator and the clock skew due to the different temperatures affecting the nodes. By applying a nonlinear filtering, the homomorphic filtering, both components are separated in an effective way. A correction factor based on temperature, which can be applied to any synchronization protocol, is proposed. For testing it, an improvement of the FTSP synchronization protocol has been developed and physically tested under temperature variation scenarios using TelosB motes flashed with the IEEE 802.15.4 implementation supplied by TinyOS. This work has been partly supported by the Computer Architecture, Electronics and Electronic Technology Department at University of Cordoba (Spain), Spanish Grants P11-TIC-7462 and DPI2013-47347-C2-2-R and European Grant EEA 014-ABELCM-2013.

Research in mobile robotics requires fully operative autonomous systems to test and compare algorithms in real-world conditions. However, the implementation of such systems remains to be a highly time-consuming process. In this work, we present an robot operating system (ROS)-based navigation framework that allows the generation of new autonomous navigation applications in a fast and simple way. Our framework provides a powerful basic structure based on abstraction levels that ease the implementation of minimal solutions with all the functionalities required to implement a whole autonomous system. This approach helps to keep the focus in any sub-problem of interest (i.g. localization or control) while permitting to carry out experimental tests in the context of a complete application. To show the validity of the proposed framework we implement an autonomous navigation system for a ground robot using a localization module that fuses global navigation satellite system (GNSS) positioning and Monte Carlo localization by means of a Kalman filter. Experimental tests are performed in two different outdoor environments, over more than twenty kilometers. All the developed software is available in a GitHub repository. This work has been supported by InterregV Sudoe and FEDER programs of European Commission through the COMMANDIA project SOE2/P1/F0638, and by the Spanish Government through the FPU grant FPU15/04446 and the research project RTI2018-094279-B-I00.

Grid-based perception techniques in the automotive sector based on fusing information from different sensors and their robust perceptions of the environment are proliferating in the industry. However, one of the main drawbacks of these techniques is the traditionally prohibitive, high computing performance that is required for embedded automotive systems. In this work, the capabilities of new computing architectures that embed these algorithms are assessed in a real car. The paper compares two ad hoc optimized designs of the Bayesian Occupancy Filter; one for General Purpose Graphics Processing Unit (GPGPU) and the other for Field-Programmable Gate Array (FPGA). The resulting implementations are compared in terms of development effort, accuracy and performance, using datasets from a realistic simulator and from a real automated vehicle., This work has been partially funded by the Spanish Ministry of Economy and Competitiveness with the National Projects TCAP-AUTO (RTC-2015-3942-4) and NAVEGASE (DPI2014-53525-C3-1-R).

In this work, we have developed a multisensor system for rehabilitation and interaction with persons with motor and cognitive disabilities. The system enables them to perform different therapies using multiple modes of interaction (head and body pose, hand gestures, voice, touch and gaze) depending on the type and degree of disability. Through a training process, the system can be customized enabling the definition of patients’ own gestures for each sensor. The system is integrated with a range of applications for rehabilitation. Examples of these applications are puzzle solving, mazes and text writing using predictive text tools. The system also provides a flexible and modular framework for the development of new applications oriented towards novel rehabilitation therapies. The proposed system has been integrated in a mobile robotic platform and uses low-cost sensors allowing to perform non-intrusive rehabilitation therapies at home. Videos showing the proposed system and users interacting in different ways (multimodal) are available on our project website www.rovit.ua.es/patente/. This work has been funded by the Spanish Government TIN2016-76515-R grant for the COMBAHO project, supported with Feder funds. This work has also been supported by a Spanish national grant for PhD studies FPU15/04516.

Graphical abstractDisplay Omitted HighlightsWe propose three improvements for those smoothing methods, improving the color quality or the computation time.One is based on entropy, speeding up the whole process.The second one obtains the optimal processing radius to improve the color quality.The last one uses a heuristic approach to select the optimal radius while improving the speed up. RGB-D sensors are capable of providing 3D points (depth) together with color information associated with each point. These sensors suffer from different sources of noise. With some kinds of RGB-D sensors, it is possible to pre-process the color image before assigning the color information to the 3D data. However, with other kinds of sensors that is not possible: RGB-D data must be processed directly. In this paper, we compare different approaches for noise and artifacts reduction: Gaussian, mean and bilateral filter. These methods are time consuming when managing 3D data, which can be a problem with several real time applications. We propose new methods to accelerate the whole process and improve the quality of the color information using entropy information. Entropy provides a framework for speeding up the involved methods allowing certain data not to be processed if the entropy value of that data is over or under a given threshold. The experimental results provide a way to balance the quality and the acceleration of these methods. The current results show that our methods improve both the image quality and processing time, as compared to the original methods.

Sensing techniques are important for solving problems of uncertainty inherent to intelligent grasping tasks. The main goal here is to present a visual sensing system based on range imaging technology for robot manipulation of non-rigid objects. Our proposal provides a suitable visual perception system of complex grasping tasks to support a robot controller when other sensor systems, such as tactile and force, are not able to obtain useful data relevant to the grasping manipulation task. In particular, a new visual approach based on RGBD data was implemented to help a robot controller carry out intelligent manipulation tasks with flexible objects. The proposed method supervises the interaction between the grasped object and the robot hand in order to avoid poor contact between the fingertips and an object when there is neither force nor pressure data. This new approach is also used to measure changes to the shape of an object’s surfaces and so allows us to find deformations caused by inappropriate pressure being applied by the hand’s fingers. Test was carried out for grasping tasks involving several flexible household objects with a multi-fingered robot hand working in real time. Our approach generates pulses from the deformation detection method and sends an event message to the robot controller when surface deformation is detected. In comparison with other methods, the obtained results reveal that our visual pipeline does not use deformations models of objects and materials, as well as the approach works well both planar and 3D household objects in real time. In addition, our method does not depend on the pose of the robot hand because the location of the reference system is computed from a recognition process of a pattern located place at the robot forearm. The presented experiments demonstrate that the proposed method accomplishes a good monitoring of grasping task with several objects and different grasping configurations in indoor environments. The research leading to these result has received funding from the Spanish Government and European FEDER funds (DPI2015-68087R), the Valencia Regional Government (PROMETEO/2013/085) as well as the pre-doctoral grant BES-2013-062864.

During grasping and intelligent robotic manipulation tasks, the camera position relative to the scene changes dramatically because the robot is moving to adapt its path and correctly grasp objects. This is because the camera is mounted at the robot effector. For this reason, in this type of environment, a visual recognition system must be implemented to recognize and “automatically and autonomously” obtain the positions of objects in the scene. Furthermore, in industrial environments, all objects that are manipulated by robots are made of the same material and cannot be differentiated by features such as texture or color. In this work, first, a study and analysis of 3D recognition descriptors has been completed for application in these environments. Second, a visual recognition system designed from specific distributed client-server architecture has been proposed to be applied in the recognition process of industrial objects without these appearance features. Our system has been implemented to overcome problems of recognition when the objects can only be recognized by geometric shape and the simplicity of shapes could create ambiguity. Finally, some real tests are performed and illustrated to verify the satisfactory performance of the proposed system. The research leading to these result has received funding from the Spanish Government and European FEDER funds (DPI2012-32390) and the Valencia Regional Government (PROMETEO/2013/085).

Industrial and service robots deal with the complex task of grasping objects that have different shapes and which are seen from diverse points of view. In order to autonomously perform grasps, the robot must calculate where to place its robotic hand to ensure that the grasp is stable. We propose a method to find the best pair of grasping points given a three-dimensional point cloud with the partial view of an unknown object. We use a set of straightforward geometric rules to explore the cloud and propose grasping points on the surface of the object. We then adapt the pair of contacts to a multi-fingered hand used in experimentation. We prove that, after performing 500 grasps of different objects, our approach is fast, taking an average of 17.5 ms to propose contacts, while attaining a grasp success rate of 85.5%. Moreover, the method is sufficiently flexible and stable to work with objects in changing environments, such as those confronted by industrial or service robots. This work was funded by the Spanish Ministry of Economy, Industry and Competitiveness through the project DPI2015-68087-R (pre-doctoral grant BES-2016-078290) as well as the European Commission and FEDER funds through the COMMANDIA project (SOE2/P1/F0638), action supported by Interreg-V Sudoe.

Finding an appropriate image representation is a crucial problem in robotics. This problem has been classically addressed by means of computer vision techniques, where local and global features are used. The selection or/and combination of different features is carried out by taking into account repeatability and distinctiveness, but also the specific problem to solve. In this article, we propose the generation of image descriptors from general purpose semantic annotations. This approach has been evaluated as source of information for a scene classifier, and specifically using Clarifai as the semantic annotation tool. The experimentation has been carried out using the ViDRILO toolbox as benchmark, which includes a comparison of state-of-the-art global features and tools to make comparisons among them. According to the experimental results, the proposed descriptor performs similarly to well-known domain-specific image descriptors based on global features in a scene classification task. Moreover, the proposed descriptor is based on generalist annotations without any type of problem-oriented parameter tuning. Finding an appropriate image representation is a crucial problem in robotics. This problem has been classically addressed by means of computer vision techniques, where local and global features are used. The selection or/and combination of different features is carried out by taking into account repeatability and distinctiveness, but also the specific problem to solve. In this article, we propose the generation of image descriptors from general purpose semantic annotations. This approach has been evaluated as source of information for a scene classifier, and specifically using Clarifai as the semantic annotation tool. The experimentation has been carried out using the ViDRILO toolbox as benchmark, which includes a comparison of state-of-the-art global features and tools to make comparisons among them. According to the experimental results, the proposed descriptor performs similarly to well-known domain-specific image descriptors based on global features in a scene classification task. Moreover, the proposed descriptor is based on generalist annotations without any type of problem-oriented parameter tuning.

The use of 3D data in mobile robotics applications provides valuable information about the robot’s environment. However usually the huge amount of 3D information is difficult to manage due to the fact that the robot storage system and computing capabilities are insufficient. Therefore, a data compression method is necessary to store and process this information while preserving as much information as possible. A few methods have been proposed to compress 3D information. Nevertheless, there does not exist a consistent public benchmark for comparing the results (compression level, distance reconstructed error, etc.) obtained with different methods. In this paper, we propose a dataset composed of a set of 3D point clouds with different structure and texture variability to evaluate the results obtained from 3D data compression methods. We also provide useful tools for comparing compression methods, using as a baseline the results obtained by existing relevant compression methods. This work was partially supported by grant DPI2013-40534-R of the Ministerio of Economia y Competitividad of the Spanish Government, supported with Feder funds, and Valencian’s Government project GV/2014/097.

Similarity functions are a fundamental component of many learning algorithms. When dealing with string or tree-structured data, measures based on the edit distance are widely used, and there exist a few methods for learning them from data. In this context, we recently proposed GESL (Bellet et al., 2012 3), an approach to string edit similarity learning based on loss minimization which offers theoretical guarantees as to the generalization ability and discriminative power of the learned similarities. In this paper, we argue that GESL, which has been originally dedicated to deal with strings, can be extended to trees and lead to powerful and competitive similarities. We illustrate this claim on a music recognition task, namely melody classification, where each piece is represented as a tree modeling its structure as well as rhythm and pitch information. The results show that GESL outperforms standard as well as probabilistically-learned edit distances and that it is able to describe consistently the underlying melodic similarity model. HighlightsNew framework for learning tree edit distances thanks to a convex optimization problem.This allows us to have strong theoretical justification without increasing the complexity.We apply our method to classify musical pieces represented by trees.Our method outperforms standard as well as probabilistically-learned edit distances.

In this work, a modified version of the elastic bunch graph matching (EBGM) algorithm for face recognition is introduced. First, faces are detected by using a fuzzy skin detector based on the RGB color space. Then, the fiducial points for the facial graph are extracted automatically by adjusting a grid of points to the result of an edge detector. After that, the position of the nodes, their relation with their neighbors and their Gabor jets are calculated in order to obtain the feature vector defining each face. A self-organizing map (SOM) framework is shown afterwards. Thus, the calculation of the winning neuron and the recognition process are performed by using a similarity function that takes into account both the geometric and texture information of the facial graph. The set of experiments carried out for our SOM-EBGM method shows the accuracy of our proposal when compared with other state-of the-art methods.

Smart Cities are called to change the daily life of human beings. This concept permits improving the efficiency of our cities in several areas such as the use of water, energy consumption, waste treatment, and mobility both for people as well as vehicles throughout the city. This represents an interconnected scenario in which thousands of embedded devices need to work in a collaborative way both for sensing and modifying the environment properly. Under this scenario, the majority of devices will use wireless protocols for communicating among them, representing a challenge for optimizing the use of the electromagnetic spectrum. When the density of deployed nodes increases, the competition for using the physical medium becomes harder and, in consequence, traffic collisions will be higher, affecting data-rates in the communication process. This work presents mDARAL, a multi-radio routing algorithm based on the Dynamic and Adaptive Radio Algorithm (DARAL), which has the capability of isolating groups of nodes into sub-networks. The nodes of each sub-network will communicate among them using a dedicated radio frequency, thus isolating the use of the radio channel to a reduced number of nodes. Each sub-network will have a master node with two physical radios, one for communicating with its neighbours and the other for being the contact point among its group and other sub-networks. The communication among sub-networks is done through master nodes in a dedicated radio frequency. The algorithm works to maximize the overall performance of the network through the distribution of the traffic messages into unoccupied frequencies. The obtained results show that mDARAL achieves great improvement in terms of the number of control messages necessary to connect a node to the network, convergence time and energy consumption during the connection phase compared to DARAL. This work has been partly supported by the Spanish grants DPI2013-47347-C2-2-R, P11-TIC-7462 and TIN2015-67020-P (Spanish “Ministerio de Economía y Competitividad” and European Regional Development Funds (ERDF)).

Many fast similarity search techniques relies on the use of pivots (specially selected points in the data set). Using these points, specific structures (indexes) are built speeding up the search when queering. Usually, pivot selection techniques are incremental, being the first one randomly chosen. This article explores several techniques to choose the first pivot in a tree-based fast similarity search technique. We provide experimental results showing that an adequate choice of this pivot leads to significant reductions in distance computations and time complexity. Moreover, most pivot tree-based indexes emphasizes in building balanced trees. We provide experimentally and theoretical support that very unbalanced trees can be a better choice than balanced ones. The authors thank the Spanish CICyT for partial support of this work through projects TIN2009-14205-C04-C1, the Ist Programme of the European Community, under the Pascal Network of Excellence, (Ist– 2006-216886), and the program Consolider Ingenio 2010 (Csd2007-00018).